Multi-label Classification: Inconsistency, Ambiguity and Class Balanced KNN Classification
Hua Wang, Chris H. Q. Ding, Heng Huang · 2010
Many existing researches employ one-vs-others approach to decompose a multi-label classification problem into a set of 2-class classification problems, one for each class. This ap-proach is valid in traditional single-label classification. How-ever, it incurs training inconsistency in multi-label classifica-tion, because a multi-label data point could belong to more than one class. In this work, we further develop classical K-Nearest Neighbor classifier and propose a novel Class Balanced K-Nearest Neighbor (BKNN) approach for multi-label classification by emphasizing balanced usage of data from all the classes. In addition, we also propose a Class Balanced Linear Discriminant Analysis approach to address high-dimensional multi-label input data. Promising exper-imental results on three broadly used multi-label data sets demonstrate the effectiveness of our approach.